Model: GELU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 771 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 8.0469. Min Range: -8.0469

Max Act: 8.0469. Min Act: -0.1700

Data Index: 922162 (C4 (Web Text))

Max Activating Token Index: 307

Text #1

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.8594. Min Act: -0.1700

Data Index: 1302524 (C4 (Web Text))

Max Activating Token Index: 978

Text #2

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.7969. Min Act: -0.1700

Data Index: 395353 (C4 (Web Text))

Max Activating Token Index: 889

Text #3

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.7969. Min Act: -0.1700

Data Index: 1187585 (C4 (Web Text))

Max Activating Token Index: 1016

Text #4

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.6719. Min Act: -0.1700

Data Index: 1216232 (C4 (Web Text))

Max Activating Token Index: 864

Text #5

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.6719. Min Act: -0.1700

Data Index: 1544528 (Python Code)

Max Activating Token Index: 891

Text #6

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.4219. Min Act: -0.1700

Data Index: 492749 (C4 (Web Text))

Max Activating Token Index: 901

Text #7

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.2969. Min Act: -0.1700

Data Index: 126173 (C4 (Web Text))

Max Activating Token Index: 923

Text #8

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.5469. Min Act: -0.1700

Data Index: 516605 (C4 (Web Text))

Max Activating Token Index: 897

Text #9

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.4844. Min Act: -0.1700

Data Index: 363247 (C4 (Web Text))

Max Activating Token Index: 451

Text #10

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.3594. Min Act: -0.1700

Data Index: 810187 (C4 (Web Text))

Max Activating Token Index: 491

Text #11

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.4844. Min Act: -0.1700

Data Index: 1024752 (C4 (Web Text))

Max Activating Token Index: 728

Text #12

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.4219. Min Act: -0.1700

Data Index: 504932 (C4 (Web Text))

Max Activating Token Index: 970

Text #13

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.3594. Min Act: -0.1700

Data Index: 230830 (C4 (Web Text))

Max Activating Token Index: 408

Text #14

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.4219. Min Act: -0.1700

Data Index: 900572 (C4 (Web Text))

Max Activating Token Index: 231

Text #15

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.4219. Min Act: -0.1700

Data Index: 1442218 (Python Code)

Max Activating Token Index: 704

Text #16

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.2969. Min Act: -0.1700

Data Index: 993558 (C4 (Web Text))

Max Activating Token Index: 325

Text #17

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.2969. Min Act: -0.1700

Data Index: 801746 (C4 (Web Text))

Max Activating Token Index: 499

Text #18

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.3594. Min Act: -0.1700

Data Index: 315159 (C4 (Web Text))

Max Activating Token Index: 486

Text #19

Max Range: 8.0469. Min Range: -8.0469

Max Act: 7.2344. Min Act: -0.1700

Data Index: 876306 (C4 (Web Text))

Max Activating Token Index: 922